The goal of this project is to obtain population pharmacokinetic parameters of antiepileptic drugs (AEDs) in elderly nursing home patients using a sophisticated software tool, NONMEM (Non Linear Mixed Effects Models). This accurate knowledge will allow proper utilization of AEDs in the elderly patient population. Though there is information in patients older than 65 years. Thus, the initial doses used and the principles of optimization are extrapolated from a younger age group. Considering the profound physiologic changes which occur during the aging process, and the high number of medications elderly patients receive, it is highly likely there are profound changes in the pharmacokinetics of AEDs in elderly individuals. If so, the entire dosing plans in use today for the large population of elderly patients with epilepsy may be less than optimal. Through the use of NONMEM, a pharmacostatistical tool, this project will obtain these parameters from routinely collected clinical data from over 50,000 elderly patients residing in nursing homes and long term care facilities. In addition, their variability and identification of important patient characteristics that help define these pharmacokinetic estimates will be determined. These results will be compared to a gender matched set of younger patients from the current program project grant NONMEM database. The primary parameter of interest for drugs with first order elimination is Clearance. Secondary parameters are Volume of Distribution and absorption rate constant. For drugs with Michaelis-Menten type elimination (eg, phenytoin), Vmax and km will be determined. The initial drugs to be analyzed will be phenytoin, carbamazepine, valproic acid, and phenobarbital. Covariates that will be tested to determine their importance in estimating these parameters are: age, weight, gender, race, concomitant medications, and concomitant diagnoses. These variables will be obtained from the Minimum Data Set (MDS 2.0) form that is required for all Medicare certified nursing homes. The population pharmacokinetic parameters and their inter- and intra-individual variability determined from this project can form the basis of dosing guidelines that can be utilized prospectively in the elderly population.